69% of Americans back forcing AI firms to hand half their stock to public fund

In June, a Verasight survey of 1,690 US adults found that 69% support forcing AI companies to transfer 50% of their stock to a public sovereign wealth fund. The finding surprised many because, as the reporting notes, an idea that sounded radical a year ago is now held by a clear majority. The survey was conducted in June, the same month Senator Bernie Sanders introduced the American AI Sovereign Wealth Fund Act, the legislative embodiment of that proposal.

Sanders pitched the fund as worth roughly $7 trillion（約1100兆円） and argued that because the public funded the underlying research and infrastructure, the public should share the returns. Verasight chief executive Benjamin Leff explained that the public sees such funds as a way to route the gains of the AI industry back to society. The framing—that this is not seizure but redistribution of publicly funded returns—appears to be persuasive. Senator Ed Markey has also signaled alignment, listing "sharing the AI wealth" among the six priorities of his recent AI Accountability Agenda.

The context driving this shift is a labor market absorbing cascading bad news. Tech accounted for close to a third of US layoffs in the first half of 2026, with AI increasingly named as the cause. The projections are grimmer: Goldman Sachs economist Joseph Briggs estimated more than 9% of the labor force—around 15 million workers—could lose jobs across a decade-long AI transition. Crucially, the same companies announcing layoffs are simultaneously raising their AI capital spending, a juxtaposition that makes the ownership argument land. The human cost of this transition is not evenly distributed, which the polling captures as resentment, though perhaps not as precision about the remedy.

Opponents raise multiple objections. Critics view the policy as forced seizure of private property dressed as redistribution and argue it would chill investment and drive AI development offshore. Sam Altman has disputed the premise itself, arguing an AI jobs apocalypse is unlikely, which would undermine a policy built on mass displacement. Pollsters also note that survey wording matters: asking whether firms should be "forced" to transfer stock invites higher support than asking about trade-offs. Other countries are experimenting with different tools—Chinese courts have ruled that replacing a worker with AI is not lawful grounds for dismissal, a protection absent in the US and EU. Sanders's bill will not pass this Congress, but the polling signals a meaningful shift: the question is no longer whether the public gets a claim on AI's upside, only what form the claim takes.

The shift in public opinion reflects a deepening tension between AI's economic promise and immediate labor-market pain. Tech layoffs in the first half of 2026 accounted for close to a third of all US job losses, even as the companies responsible announced record AI capital spending. That stark contrast—job cuts paired with massive investment—has made the Sanders proposal's framing (the public funded the research, so the public should share the gains) stick in a way it might not have a year earlier.

The policy itself remains unlikely to become law this Congress, and the arguments against it are substantive: equity seizure could deter investment, and if Sam Altman and others are right that mass job displacement is not inevitable, the premise collapses. The polling evidence also carries methodological caveats—asking whether firms should be "forced" to transfer stock may overstate support compared to asking about concrete trade-offs.

Yet the poll's broader signal is significant. The debate has shifted from whether AI will displace workers to how the gains should be distributed. Senator Ed Markey's recent AI Accountability Agenda lists "sharing the AI wealth" among its priorities, suggesting the ownership question is now mainstream political territory. That Overton window shift—from denial to distribution—may matter more than any single bill.